The Photosystem II (PSII) reaction center contains two homologous polypeptides, D1 and D2, that are responsible for the coordination of the primary photoreactants Satoh, K. (1993) in The Photosynthetic Reaction Center (Deisenhofer, J. and Norris, J. R., eds.) Vol. I, pp. 289-318, Academic Press, New York; Seibert, M. (1993) in The Photosynthetic Reaction Center (Deisenhofer, J. and Norris, J. R., eds.) Vol. I, pp. 319-356, Academic Press, New York!. The D1 polypeptide of PSII is present in all organisms that use oxygenic photosynthesis to fuel metabolism. The source of electrons for the electron transport chain of oxygenic photosynthetic organisms is water. The oxidation of water to molecular oxygen occurs on a tetranuclear manganese cluster that is thought to be associated with the D1 polypeptide Nixon et al., Biochem. 31, 10859-10871 (1992)!.
D1 is expressed in precursor form Grebanier et al., J. Cell Biol., 78, 734, (1978); Reisfeld et al., Eur. J. Biochem., 124, 125, (1982); Minami, J. B. and Watanabe, A. Plant. Cell Physiol. 26, 839-846 (1985)!, inserted into the thylakoid membrane, and processed at its C-terminus Marder et al., J. Biol. Chem. 259, 3900-3908 (1984); Diner et al., J. Biol Chem. 263, 8972-8980 (1988); Nixon et al., Biochem. 31, 10859-10871 (1992)!. In cyanobacteria 16 residues are cleaved from precursor D1 (pre-D1) Nixon et al., (1992), supra! whereas 9 residues are cleaved in higher plants (Takahashi et al., FEBS Lett. 240, 6-8 (1988); Takahashi et al., Plant Cell Physiol. 31, 273-280 (1990)), with processing occurring in all at the carboxy side of D1-Ala344. It has been suggested that this processing is effected by a protease enzyme, D1 protease.
Failure to process the carboxy-terminal extension of pre-D1 results in the inability to fully assemble the manganese cluster necessary for photosynthetic water oxidation Diner et al., J. Biol Chem. 263, 8972-8980 (1988); Taylor et al., FEBS Lett. 235, 109-116 (1988)!. As the oxidation of water is absolutely essential to photosynthesis, prevention of this process prevents photoautotrophic growth of all cyanobacteria, algae and higher plants. Agents that inhibit the C-terminal processing of the D1 protein represent herbicidal candidates.
Although several proteins termed "D1 protease" as well as genes ostensibly encoding D1 protease enzymes have been isolated from cyanobacteria, algae, and higher plants, there is no evidence until now that these enzymes are responsible for in vivo C-terminal processing of the D1 polypeptide. For example, Shestakov et al. J. Biol Chem. 269, 19354-19359 (1994)! and Anbudurai et al. Proc. Natl. Acad. Sci., USA 91, 8082-8086 (1994)! teach the isolation of the ctpA gene from the cyanobacterium Synechocystis, a mutation which impairs the C-terminal processing of the pre-D1. The characterization of this gene as encoding a D1 protease was made on the basis of the impairment, measured in vivo of pre-D 1 processing in vivo and not on the basis of enzyme activity since no protein associated with this gene has as yet been isolated. Further studies by Applicants have shown, however, that the inactivation of the ctpA gene does not completely remove the ability of the mutant strain to process D1, suggesting that this protein is not wholly responsible for D1 processing.
An enzyme demonstrating D1 protease activity in vitro has been isolated from spinach Fujita et al., Plant Cell Physiol. 36(7) 1169-1177 (1995)! and the gene encoding the enzyme has been cloned and sequenced Inagaki et al., Plant Mol. Biol., 30(1), 39-50 (1996)!. In vitro assays have shown that the spinach enzyme is capable of using a C-terminal fragment of the pre-D1 protein (consisting of 24 amino acids) as a substrate, but there has been no demonstration of a link between this enzyme and the in vivo processing of the pre-D1 protein Taguchi et al., J. Biol. Chem., 270(18), 10711-16 (1995)!.
Pre-D1 protein processing activities have been isolated and partially purified from Scenedesmus and Pisum, Packer et al., Curr. Res. Photosynth., Proc. Int. Conf. Photosynth., 8th (1990), Meeting Date 1989, Volume 3, 759-62. Editor(s): Baltscheffsky, Margareta. Publisher: Kluwer, Dordrecht, Neth.! and from maize Magnin et al., Res. Photosynth., Proc. Int. Congr. Photosynth., 9th (1992), Volume 2, 211-14. Editor(s): Murata, Norio. Publisher: Kluwer, Dordrecht, Neth.!. These enzymes demonstrated activity in an in vitro PSII particle assay; however, no demonstration of in vivo activity of these enzymes has been obtained until now.
Isolation of an enzyme from a plant that has pre-D1 processing activity is not de facto evidence that it is indeed responsible for in vivo pre-D1 protein processing. For example, an enzyme contained in periplasmic lysates of E. coli tail-specific protease has been identified Silber et al., Proc. Natl. Acad. Sci., USA 89, 295-299 (1992); Hara et al., J. Bacteriol. 173, 4799-4813 (1991)! that has about 30% identity to the putative D1 protease. Further, an enzymatic activity has been isolated by the Applicants from periplasmic isolates which has pre-D1 protein processing activity in vitro. While it is probable that these are one and the same enzyme, E. coli does not contain D1 and does not perform oxygenic photo-synthesis. Therefore, it cannot be concluded that an enzyme is D1 protease purely on the basis of its homology to known D1 protease-encoding genes and evidence of in vitro activity.
Thus, in order to develop a method for the screening of herbicidal agents that target D1 protease, one problem to be solved is to positively identify the enzyme that is responsible for in vivo processing of the pre-D1 protein.
Methods for assaying the presence of pre-D1 protein processing activity are known. For example, Hunt et al., Res. Photosynth., Proc. Int. Congr. Photosynth., 9th (1992), Volume 2, 207-10. Editor(s): Murata, Norio. Publisher: Kluwer, Dordrecht, Neth.! teach an assay system using a truncated peptide substrate based on the C-terminal region of the D1 protein. Similarly, Packer et al., (Curr. Res. Photosynth., Proc. Int. Conf. Photosynth., supra) describe an assay using PSII thylakoid particles from the Scendedesmus D1 protease-deficient mutant LF-1. LF-1 PSII particles are incubated with a solution extracted from sonicated wildtype Scenedesmus thylakoids and D1 is processed to its normal mature size where the Mn complex is then photoligated and the photooxidation of water is detected. Finally, Taguchi et al., J. Biol Chem., 270(18), 10711-16 (1995)! teach an assay method using purified spinach D1 protease and either in vitro truncated D1 protein or synthetic oligopeptides, both containing the D1 C-terminus as a substrate. Enzyme products are identified by gel shift analysis and HPLC, respectively.
Although assay methods such as these are useful for the detection of pre-D1 processing activity, they are not readily adaptable for commercially useful high throughput screens because they use large quantities of enzyme, rely on identification of enzyme substrate by either gel or HPLC analyis, and take hours to give results. Additionally, assays using truncated D1 as substrates (Hunt et al., supra; Taguchi et al., supra) must run the assay at a pH higher than that at which the enzyme functions in vivo.
Another problem to be solved then is to develop an assay system that is facile and adaptable to high through-put screening.